Automatic Density Peaks Clustering Using DNA Genetic Algorithm Optimized Data Field and Gaussian Process

Abstract

Clustering by fast search and finding of Density Peaks (called as DPC) introduced by Alex Rodríguez and Alessandro Laio attracted much attention in the field of pattern recognition and artificial intelligence. However, DPC still has a lot of defects that are not resolved. Firstly, the local density [Formula: see text] of point [Formula: see text] is affected by the cutoff distance [Formula: see text], which can influence the clustering result, especially for small real-world cases. Secondly, the number of clusters is still found intuitively by using the decision diagram to select the cluster centers. In order to overcome these defects, this paper proposes an automatic density peaks clustering approach using DNA genetic algorithm optimized data field and Gaussian process (referred to as ADPC-DNAGA). ADPC-DNAGA can extract the optimal value of threshold with the potential entropy of data field and automatically determine the cluster centers by Gaussian method. For any data set to be clustered, the threshold can be calculated from the data set objectively rather than the empirical estimation. The proposed clustering algorithm is benchmarked on publicly available synthetic and real-world datasets which are commonly used for testing the performance of clustering algorithms. The clustering results are compared not only with that of DPC but also with that of several well-known clustering algorithms such as Affinity Propagation, DBSCAN and Spectral Cluster. The experimental results demonstrate that our proposed clustering algorithm can find the optimal cutoff distance [Formula: see text], to automatically identify clusters, regardless of their shape and dimension of the embedded space, and can often outperform the comparisons.

Adaptive Density Peaks Clustering Based on K-Nearest Neighbor and Gini Coefficient

“…From the results of testing DPC on several different data sets, it is clear that DPC can perform well in many instances, though the following drawbacks are apparent: [32][33][34][35][36][37][38][39][40][41] i) After the distance matrix is constructed, two important variables i  and i  need to be solved, and the solution of these two parameters is closely related to the only parameter c d of this algorithm. The selection of the distance threshold is directly related to the quality of the clustering effect, especially on some small and mediumsized data sets, the density peaks clustering is sensitive to c d anomaly.…”

HaloDPC: An Improved Recognition Method on Halo Node for Density Peak Clustering Algorithm

“…7,10,23,27 Traditional methods of clustering can be broadly categorized into those of hierarchical, partitioning, density-based, model-based, grid-based, and softcomputing. 17,19,21,25 Inspired by DBSCAN, 6 many density-based clustering methods 5,6,20 have been proposed. In the last decade, DBSCAN has had a big impact on the data mining research community due to its capability of discovering clusters with arbitrary shapes and noise detection.…”

“…18 Both the idea of local density maxima from mean-shift 3 and the idea of only one parameter of the distance between data points from K-Medoids 15 are adopted by DPC. Focusing on this method, several researches 1,4,[11][12][13]22,25,26 have been carried out for improving its capabilities.…”

A Kernel-Based Intuitionistic Fuzzy C-Means Clustering Using a DNA Genetic Algorithm for Magnetic Resonance Image Segmentation

“…The DNA genetic algorithm, based on DNA computing [22] and the Genetic Algorithm (GA) [23], have been recently introduced to solve complex optimization problems in many areas, such as, chemical engineering process parameter estimation [24], function optimization [25], clustering analysis [26,27], and membrane computation [28]. This technique can be used to solve the aforementioned optimization problem.…”